diff --git a/checkin_notes b/checkin_notes
index 4f9827822d..3ef9fae45f 100644
--- a/checkin_notes
+++ b/checkin_notes
@@ -4164,3 +4164,14 @@ David  24 Apr 2009
         work_fetch.cpp
     html/ops/
         team_import.php
+
+David  24 Apr 2009
+    - client: tweak to 4/21 checkin.
+        After finding the "most capable" GPU,
+        ignore FLOPS in deciding what GPUs are equivalent to it.
+        This opens up the possibility that the client will get jobs
+        that it won't be able to finish in time.
+        But it still avoids getting jobs that will crash.
+
+    lib/
+        coproc.cpp
diff --git a/lib/coproc.cpp b/lib/coproc.cpp
index c509dd6a1c..afe10d6634 100644
--- a/lib/coproc.cpp
+++ b/lib/coproc.cpp
@@ -144,7 +144,7 @@ COPROC* COPROCS::lookup(const char* type) {
 
 // return 1/-1/0 if device 1 is more/less/same capable than device 2
 //
-int cuda_compare(COPROC_CUDA& c1, COPROC_CUDA& c2) {
+int cuda_compare(COPROC_CUDA& c1, COPROC_CUDA& c2, bool ignore_flops) {
     if (c1.prop.major > c2.prop.major) return 1;
     if (c1.prop.major < c2.prop.major) return -1;
     if (c1.prop.minor > c2.prop.minor) return 1;
@@ -153,6 +153,7 @@ int cuda_compare(COPROC_CUDA& c1, COPROC_CUDA& c2) {
     if (c1.drvVersion < c2.drvVersion) return -1; 
     if (c1.prop.totalGlobalMem > c2.prop.totalGlobalMem) return 1;
     if (c1.prop.totalGlobalMem < c2.prop.totalGlobalMem) return -1;
+    if (ignore_flops) return 0;
 	double s1 = c1.flops_estimate();
 	double s2 = c1.flops_estimate();
 	if (s1 > s2) return 1;
@@ -269,7 +270,7 @@ void COPROC_CUDA::get(COPROCS& coprocs, vector<string>& strings) {
     for (i=0; i<gpus.size(); i++) {
         if (i==0) {
             best = gpus[i];
-        } else if (cuda_compare(gpus[i], best) > 0) {
+        } else if (cuda_compare(gpus[i], best, false) > 0) {
             best = gpus[i];
         }
     }
@@ -281,7 +282,7 @@ void COPROC_CUDA::get(COPROCS& coprocs, vector<string>& strings) {
     for (i=0; i<gpus.size(); i++) {
         char buf[256];
 		cc.description(buf);
-        if (!cuda_compare(gpus[i], best)) {
+        if (!cuda_compare(gpus[i], best, true)) {
             best.device_nums[best.count] = gpus[i].device_num;
             best.count++;
             strings.push_back("CUDA device: "+string(buf));